Loop issue with PVST+

[da-alb]

Hi everyone,

I have this network diagram here and I can't figure out why when I connect the link between C101 and C109 I have lots of packet loss and issues on VL60.
Every port that connects the switches has every vlan (1,10,20,30,40,60,70,80,90,100,200,300,500) tagged. I'm using RPVST+ in every cisco and the C101 (the core switch is a stack) has 8K as priority. 
The vlans are: 10,20,30,40,60,70,80,90,100. 
The blue switches are Cisco Catalyst 1300 and the green switches are HPE OfficeConnect 1820.

Any idea? What am I missing?

[da-alb]

From what I've read if an access switch like our HPE have only an uplink, configuring STP on them is not really needed. Am I wrong?

[Joseph W. Doherty]

From what I've read if an access switch like our HPE have only an uplink, configuring STP on them is not really needed. Am I wrong?

Depends on what's possible regarding accidentally creating a L2 loop and your risk tolerance.

Something like STP is really needed for by design L2 loops, such as the core ring, for redundancy, design.

As usually, switches, even L3 switches, have multiple L2 ports in the same L2 domain, something like STP avoids the impact of an accidental loop creation.

Ditto for multiple switches sharing the same L2 domain.

Your HP and Catalyst switches, I understand, share L2 domains, but if there's no way to create a second inter switch link, they don't need to share STP, but STP still highly valuable for handling accidental L2 loop formation on just the HP switch.

Does the forgoing help?

[da-alb]

I'm using rapid-pvst though

[da-alb]

We can close the thread, the issue has been resolved. It was ICMP Snooping enabled on a TP-LINK switch on VLAN 60 that flooded the network.

[MHM Cisco World]

I still think issue not solve' 

Flooding with good STP not make network down 

You have issue with STP

MHM

[pieterh]

good work,
thank you for reporting the found source of your issue